Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
Chapter One
Democracy, democratic ideals and voting
Voting is at the very heart of the democratic process. Very few people in the modern world believe in the myth of the statesman/woman who, “as the mystical representative of the nation”, can discern what the people think by examining his/her own heart. If the people are to rule, there must be institutionally defined channels of participation. There are many possible forms of popular participation. People can join political parties, they can inform themselves on political issues, they can discuss and debate these issues with their friends, but for most of us, formal participation in the exercise of political power comes down to the solitary act of voting. There are many factors that go to make up a particular voting system. One aim of this book will be to explain what these factors are and how they can be combined. It would probably be a safe guess that most voters would have some basic accurate knowledge of their own voting system in terms of how to fill out a valid ballot and, at least roughly, how their votes go to determine the results. But it would probably also be safe to say the following things: under normal circumstances most people take the voting systems that they are used to for granted. There are many technical aspects of such systems concerning which many people will have only the vaguest idea. The historical origins and rationale may not be fully understood. Furthermore, citizens of one state may lack detailed knowledge of the voting systems of other states and will probably not have engaged in any serious comparative assessment of different types of voting system and their underlying rationales. This relatively uncritical complacency with what people are accustomed to can, however, be seriously disturbed when the circumstances are not normal. A case in point was the U.S. presidential election of November 2000. The basis of the problem in that instance lay in the closeness of the contest between George W. Bush and Al Gore. Events began to unfold dramatically when the major TV networks confidently predicted a Gore victory, only to retract that prediction when it became evident that victory in the electoral college depended crucially on who won the state of Florida, and that Bush seemed to be winning, but by such a small margin that recounts would be needed. The contentiousness, which in the specific case of Florida centred on such things as the late tally of postal votes, the ambiguity of the ballot paper, and
1 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
margins of error in vote counting, was compounded by the fact that even if Bush won the state of Florida, and hence the election as a whole, Gore had won a majority of the popular vote. 1 While there are, of course, serious questions concerning the democratic nature of electoral systems where there can be some degree of mismatch between the results and the popular vote, it is our contention that even more troublesome questions arise if we engage in a slightly more detailed analysis of the technicalities of voting systems, and a more searching investigation of what we might mean by “the will of the people”. To illustrate in a preliminary way the major importance of these issues, consider the following constructed electoral scenario, which though imaginary, is not very different from the U.S. presidential election of 2000 in terms of candidate support. In accordance with Mr. Morris’s prescription we will imagine an electoral system in which the result is directly dependent on the popular vote. Labelling our three candidates B, G and N, suppose that, on a turnout of 100,000,000, the percentage support for each candidate was: B G N 47% 45% 8%
Table 1
Clearly, using the vote-counting system usually referred to as the plurality system (popularly called “first past the post”), B would win by a margin of two percentage points over G, with N a very distant third. Although two percentage points over a nearest rival is a relatively narrow margin, in terms of the postulated turnout it represents a very substantial 2,000,000 votes. Presumably this, according to Mr. Morris, would be democracy, because the will of the people is being heard and heeded. The
1 By some, this was thought to place democracy in America on trial. Dick Morris, chief electoral strategist for Bill Clinton’s 1996 campaign, wrote at the time of the 2000 election: “I live in a democracy in the United States. Maybe. It is an absolute outrage for George W. Bush to become president if, as current trends suggest, he loses the popular vote by 250,000….if current vote totals stand up, Mr. Bush will win the presidency even though more people voted for Al Gore. This is not democratic….there is a clear and compelling need for the will of the people to be heard and heeded.” (Irish Times, November 10 th , 2000)
2 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
“maybe” could be erased and Americans could rest assured that they do indeed live in a democracy. The people of the Republic of Ireland also, so they believe, live in a democracy, with, according to Mr. Morris’s criterion, no “maybe” about it; the system for electing their president being already one in which the result is determined directly by the popular vote. In Ireland, however, the voting system – more strictly the balloting system and the procedures used for determining the results on the basis of the ballots – differs significantly from the simple plurality or “first past the post” system used in our first example. It is usually referred to as the Alternative Vote (AV) system and is rather more complex. To begin with, instead of being asked simply to nominate their first choice candidate, voters are invited to rank candidates according to their order of preference. A ballot form in an election with our three candidates B, G and N would look like this.
B
G
N
A voter wishing to select N before G, and G before B, would complete the ballot as follows:
B 3
G 2
N 1
The numbers 1, 2 and 3 indicate first, second and third preference. The counting of votes takes pace in several stages. At the first stage each candidate’s first preference votes are summed. With identical preferences to those in our first example, the first count would of course produce the following results:
B G N 47% 45% 8%
3 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
(47,000,000) (45,000,000) (8,000,000)
Table 2
If, as is here the case, no candidate has an overall majority (i.e. no candidate has more than 50% of the vote and hence no candidate has a total vote greater than the combined first preference votes of the other candidates) a second stage of counting takes place. Without going into the finer points, this stage begins with the elimination of the candidate with the lowest number of first preference votes; in our example this would be N. What happens next is that each ballot with N as first choice is examined to see who is the second choice. These second preference votes are totalled for each remaining candidate and added to their first preference total. Taking N’s 8,000,000 ballots, suppose that 6,000,000 identified G as second preference, while only 2,000,000 identified B. The results of the second count would look like this:
B G 49,000,000 51,000,000 (47,000,000 + (45,000,000 + 2,000,000) 6,000,000)
Table 3
In a three-candidate election, this would constitute the final result, with G winning this time, with a straight 2,000,000 majority. With absolutely identical popular preferences over the candidates, and hence, one would think, the self-same “will of the people” contradictory results are produced. How an identical “will of the people” can produce totally contradictory results is a little puzzling and definitely democratically troublesome. But perhaps all is not lost. After all, the margin between the two front- runners is only two percentage points, and the two voting systems end up selecting one of the front-runners. Perhaps all we are dealing with here is a situation in which two reasonably democratic voting procedures, not being as absolutely fine-tuned as they could be, produce contrary results when the different between the front-runners is very small. We will be able to show in detail later how the narrowness of the margin between the front-runners is not the critical factor, but this point can be demonstrated
4 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
immediately and dramatically by introducing a third type of voting system, on that could claim to have just as good democratic credentials as the two already considered. The system in question is known as the Borda Count system, named after its 18 th - century proponent, the French academician Jean Charles de Borda. Many eminent theorists have strongly argued its high democratic credentials (see, for example, M. Dummett, Voting Procedures, pp.xxx). As with the AV system, voters are invited to rank order candidates, but unlike that system there is no sequential elimination of lower- ranked candidates. Rather, each candidate is awarded a number of points for each preference ranking, the usual method being that with n being the number of candidates, n-1 points are awarded for a first preference ranking, n-2 for a second preference and so on. In our three-candidate scenario, n-1 would be 2, so a first preference vote would earn 2 points and a second preference vote would earn 1 point, with no points awarded for a third preference. In examining the operations of the AV system we postulated that of the 8,000,000 voters ranking N first, 6,000,000 rank G second, with 2,000,000 ranking B second. To analyse how the Borda count might operate, we need to construct a possible full ranking for those voters who give B and G their first preferences. Suppose, for example, that many B and G supporters are very partisan, each group preferring the third candidate to the representative of the main rival party. A possible set of preferences might then be:
B G First preferences 47,000,000 45,000,000
Second N G N B preferences 44,000,000 3,000,000 42,000,000 3,000,000
Third preferences G N B N 3,000,000 44,000,000 3,000,000 42,000,000
Table 4
Adding the exact same preference ratings for the 8 million N supporters as in the AV example, we would get the following table:
5 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
B B G G N N
1st 44m 3m 42m 3m 6m 2m
N G N B G B
2nd 44m 3m 42m 3m 6m 2m
G N B N B G
3rd 44m 3m 42m 3m 6m 2m
Table 5
The Borda count points awarded to the candidates for first preferences would then be:
B 94m (47m * 2)
G 90m (45m * 2)
N 16m (8m * 2)
with B being ahead at this stage, but what happens when the second preference points are added? Well, G would gain 6m from the second preferences of those who put N first, and B would gain 2m, resulting in a tie between B and G, as follows:
B 96m (94m + 2m)
G 96m (90m + 6m)
Now this might seem to confirm the suggestion that we are dealing with voting systems which are all credibly democratic but which are not fine-tuned enough to give the same result with a narrow margin of 2 percentage points between the front-runners. The plurality system selects B, because N’s second preferences are not taken into account at all; AV selects G, because N’s second preferences are, when transferred, given the weight of full votes; whereas in the Borda system they are given one point less than a first preference. But, it could be claimed, the different systems do select one of the front-runners. This however, is not true. We have not yet looked at how N stands when
6 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
the second preferences of B and G are counted. N was ranked second by 44 million of B supporters and 42 million of G supporters. This gives N a total of 86 million points from second preferences, which along with N’s 16 million first preference points, gives N a winning total of 102 million, using the Borda system. So, which of the three results represents “the will of the people”, which system is “democratic”, or at least “more democratic” than the others? The first response to these questions has to be a clear realisation that the result of a vote is determined by two things; the expressed preferences of voters for the options being decided upon and the system of processing those expressed preferences. Secondly, a conclusion concerning which system manages best to express “the democratic will of the people” cannot be safely based on a subjective intuition as to what we might mean by that phrase. A detailed analysis is required of how voting and electoral systems operate, both in the abstract and in the context of real, live politics. Thirdly, a comparative assessment of these systems must be made in the light of an articulated and reasoned account of what we might mean by democracy, and what goals and ideals we might be seeking to achieve in our democratic systems of government. The aim of this first chapter is to identify the democratic criteria for such a comparative assessment by examining the nature of democracy and the central values a democracy should try to achieve.
DEFINING DEMOCRACY
The task of defining democracy is not as simple and uncontentious as it might at first appear. To begin with, even a cursory perusal of the literature reveals a bewildering array of very different definitions, ranging from C.B. McPherson’s “egalitarian maximisation of human powers ( Democratic Theory: Essay on retrieval ) to Lord Bryce’s “rule by the will of the qualified majority”( quote). Despite the near-universal acceptance of “democratic” values, theorists such as R.A. Dahl claim that “democracy” literally almost means nothing, being “not so much a term of restricted and specific meaning as a vague endorsement of a popular idea”. (R.A. Dahl, Democracy and its Critics , p.2). In addition, many theorists follow W.B. Gallie in thinking that this indeterminacy and definitional multiplicity cannot be avoided, since “democracy is an essentially contested concept concerning which there will always be unresolvable definitional disagreement”. The idea is that, unlike a word like “triangle”, that is
7 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
authoritatively allocated a precise meaning in the explicit construction of an axiomatically postulated system of geometry, “democracy” enters popular usage as referring fairly indeterminately to that which those people are in favour of, who are opposed to such things as the divine right of kings, or autocratic absolute monarchy. There may be some basic negative ground rule, such as the rejection of the right of a minority to rule over everyone else, but after that the democratic tradition (or, to be precise, traditions) encompasses so many strands, some stressing one element, some others, with there being no authority to identify once and for all the absolutely essential defining elements. Take, for example, the formula of “political equality” which most, though not all, theorists of democracy will cite in their definitions. How are we to interpret the relevant type of equality? Is democracy to be identified with equality of constitutional rights of participation in an electoral representative system or should we insist that real democracy requires substantive equality of opportunity or even equality of outcome in the exercise of political power? The point is that there may not be a definitive answer to these questions; different answers can be given depending on such things as one’s ideological orientation, one’s understanding of what is possible in the organisation of the government of modern states, one’s estimate of the political capacities of the ordinary person or what one thinks are the really important values to be achieved by “democracy”. The definitional multiplicity should not, however, be the cause of despair or lead to the abandonment of the terminology. The language of democracy is so central to contemporary political discourse as to be unavoidable. The contestation of the real meaning of democracy should be seen for what it is, a polemical battle to be entered through serious argumentation concerning the possibilities and ideals to be striven for in the context of contemporary society. There are many dimensions along which the concept of democracy is contested, but there is one in particular that is critically relevant to our present enterprise, giving rise to the minimalist competitive elite model of democracy as opposed to a more robust conception of democracy as genuine effective equality among citizens in the collective government of their lives. In fact, acceptance of the competitive elite model would, for reasons to be explained shortly, so undermine the relevance of attempting our comparative evaluation of voting systems as to render it pointless. Consequently, we need to be clear from the start why we are rejecting the competitive elite model as,
8 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
allegedly, the definitive account of the meaning of democracy. Although some of the considerations underlying the appeal of the competitive elite model can be traced to the political writings of Benjamin Constant in the early 19 th century, its contemporary incarnation emerged in Schumpeter’s response to the classical elite theorists such as Pareto, Mosca and Michels (see G. Parry, Political Elites). The central contention of these theorists was that representative systems of government emerging in the developed states of their day (late 19 th to early 20 th century) could in no way be thought of as democracies. Such representative systems were simply the latest façade behind which the non-accountable power of a tiny elite was, barely, concealed. Michels in particular argued that modern representative systems were inconceivable without political parties and that political parties were so oligarchically structured that real power lay in the hands of a small elite of political activists with respect to whom the masses were, and always would be, passive subjects. Schumpeter’s reaction to this thesis was not to reject it, but to claim that to think that it undermined the possibility of democracy in the modern world was to misinterpret completely the modern meaning of democracy. He conceded that there was a classical conception of democracy to be found in writers such as Rousseau that alluded to the, for Schumpeter, incoherent notion of the popular will. But what is meant by “democracy” in the modern world is something completely different. It is not a system in which the people as a whole rule themselves rather, it is simply a method of electorally selecting rulers; in his famous definition, “democracy” “is that institutional arrangement for arriving at political decisions in which individuals acquire power to decide by means of a competitive struggle for the people’s vote” (CSD p. 269) And that, for Schumpeter, is that. So long as governments come and go on the basis of regular non-fraudulent elections, we have what in the modern world we mean by “democracy”. To be slightly more explicit, the “electorate”, for example, does not have to be based on anything approaching universal suffrage, just so long as its exclusionary principles, no matter how restrictive, conform to the dominant political culture of the society. The electorate’s selection does not need to be interpreted as a method of indirect rule in which, for example, the interests of different groups become represented. Again, quite explicitly, Schumpeter rejects arguments for more proportional electoral
9 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
systems on the grounds that such arguments simply misunderstand the role of the people in a “democracy”; their role is to cast their votes, on the basis of which a government is appointed, whose role it is to rule. Apart from the fact that the competitive elite model does encapsulate what passes for democracy in the modern world, three main arguments are used to urge the acceptance of this revised definition. The first is based on a set of considerations that can, as mentioned above, be traced back to the writings of Benjamin Constant, in particular his famous 18xx lecture comparing “the liberty of the ancients” to “the liberty of the moderns”. (constant ref.). There are three inter-related strands in Constant’s argument, the main burden of which is that the kind of “liberty” we value in the modern world is negative freedom from tyrannical government rather than the positive freedom to participate in a meaningful way in the exercise of collective political power. Firstly, deriving from the sheer size of the populations of modern states, political equality can confer on each citizen such an “imperceptible” amount of power as not to constitute any realistically valuable input into the decision-making process, in terms of either instrumental or intrinsic worth. Secondly, politics in the modern world is concerned primarily with the technical administration of “externals”; as Constant nicely put it, we want governments to provide roads that enable us to travel, but not to determine our destinations for us. Because of factors such as these, our membership of a political community as such is not at the core of our lives in the way in which it might have been for, say, a citizen of ancient Athens. Our values (in Constant’s imagery, our destinations) we want to be free to determine for ourselves. The conclusion to be drawn from these considerations might be that if electoral appointment to government institutionalises the belief that no-one has ownership of such office by right, and if the system allows for the eviction of incumbent government personnel were they to become incompetent, corrupt, tyrannical, or completely out of touch with opinions of important sections of the population, we should not be overly concerned with the technicalities of how some particular voting system works. Our negative freedom to live our lives as we individually decide would be secured. A second type of argument, associated particularly with Schumpeter, also emphasises the professionalisation of politics in the modern state. The conclusion this time, however, is that such professionalisation requires professional expertise which could
10 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
only ever be the preserve of a minority elite, far beyond the competence of the masses, whose thinking on political matters would always remain vague and indeterminate, based on a type of thinking that in other areas of life would be seen as “infantile”, merely “associative and affective” (CSD, pp. 256-264). Consequently, electoral systems should not be seen as mechanisms for the determination of “the will of the people”, a phrase that in Schumpeter’s view refers, anyway, to nothing coherent and definite. They are merely, once again, safety valves via which governments can, if necessary, be evicted. A final set of arguments also relates to the rejection of the idea that voting systems could be seen as expressions of “the popular will”, the scepticism here, however, deriving from certain complex technicalities in the way that voting systems function. Discussion of these technicalities will have to be postponed to later chapters, but the general idea developed specifically by William Riker in his Liberalism against Populism is that, though how people cast their votes will, given the rules of the system, determine the result to be implemented, we can never be sure that that result represents a uniquely determined will of the people in the issue area in question. Hence, the whole attempt to fine-tune voting systems in an effort to achieve the best approximation to a synoptic picture of the popular will has to be misguided. It is undeniable that these arguments raise serious questions about the meaning and possibility of democracy in the modern world. There are, however, reasons for resisting Schumpeter’s radical revision of the concept to the minimalist understanding of democracy as exhausted by the idea of electoral competition between elites. In the first place, we would claim that, as a matter of fact, we still do have a more generic conception of democracy, that we can analyse and investigate in abstraction from the specific exigencies of electoral competition on the scale of the nation state. It is still, for example, perfectly meaningful to imagine a group of people much smaller than the political community of a nation state, exploring in that context the ideal requirements for a maximally democratic form of collective decision-making. Our claim would be that denying ourselves this possibility by definitional fiat would risk serious neglect of certain features of collective decision-making that have been both theoretically and practically central to the democratic tradition. Furthermore, the mere fact that a more robust conception of the democratic ideal faces serious challenges to its
11 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
realisation in contemporary political systems would not in itself be sufficient reason for abandoning it as an ideal. (see Duncan and Lukes, D. Beetham e.g.). There is much point to the maintenance of a concept of democracy that would provide us with a critical perspective on the far from ideal realities of modern politics. Finally, although as we have admitted, the kinds of argument put forward for a Schumpeterian revisionism are important, they are not without their weaknesses. Take, for example, Constant’s arguments concerning the impossibility of equal power in the modern state and the changed nature of political decision-making. Constant himself, at the end of that famous lecture, greatly qualifies his conclusions, admitting that not all political decisions are merely matters of technical administration; many issues in modern politics concern what he calls “our sacred interests” (put in ref.). he even adds that self-absorption in our private affairs to the neglect of the intrinsic worth of involvement in the public life of the political community would pose a major danger to the maintenance of our political liberties as such. Turning to Schumpeter, there is undoubtedly truth in his “realistic” assessment of the actual competence of the ordinary citizen and the (very limited) time and attention that most people would be prepared to devote to serious political involvement. It would be unsafe, however, to treat present- day levels of achievement as defining the absolute limits of human possibility. Finally, while a full assessment of Riker’s arguments will have to await a more detailed technical analysis, we will claim that his almost complete scepticism concerning what can be learned of popular preferences by reference to the results of voting is much too pessimistic, and that his arguments are not in themselves reasons for abandoning the more robust conception of democracy or failing to engage in a serious comparative democratic analysis of various forms of voting and electoral system. Our claim is that a more normatively defensible and theoretically more unified and enlightening understanding can be based on a more robust conception of democracy as full, effective political equality. And it is to an elaboration of that conception that we will now turn. Although, given the contestation between conceptions of democracy, not even so famous a political philosopher as Rousseau can be cited as an authoritative source, it is interesting to begin outlining our fuller conception of democracy with what must be the pithiest definition in the literature: Rousseau’s description of democracy as the identity of sovereign and subject . The claim here is that wherever people are subject to political
12 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
power, that power is democratically organised to the extent that all those subject to the power are equal participants in its exercise. More formally and explicitly, we can define democracy as consisting in everyone having institutionally inscribed, fully effective rights of equal participation in the making of authoritatively binding decisions to which they are subject. The precise meaning of this definition can be elucidated by considering the following four contrasts. Firstly, although there is a close relationship between egalitarian values in general and a commitment to democracy in particular, they are not identical, the equality at the heart of democracy being an equality of power rather than a general equality of such things as economic wealth. Secondly, the relevant equality of power in question is not concerned with any potential differential power between individuals; democracy might indirectly depend upon economic status or the distribution of various sources of power and influence, but only in so far as these affect equality of political power as such, equality, that is, in the determination of the authoritatively binding decisions to which people are subject. Thirdly, for this equality in specifically political power to be unproblematically democratic, it must be routinised and institutionally inscribed. The power of the people to get its way by taking to the streets is a real popular power but it is not paradigmatically democratic because it is ad hoc rather than institutionally guaranteed. Regularised procedures for the guaranteed effectiveness of popular preferences in determining political outcomes are essential to normal democratic procedure. Finally, not only must there be a set of institutions in place that confer equal rights of participation, but people’s ability to utilise those rights must be equally effective, and hence whatever resources are required for such effective use must be distributed in such a way as to guarantee it. This is, admittedly an idealised maximal conception of democracy that is not likely to be realised in any real-world political community, no matter how favourable the background conditions. Despite this, there are two major reasons for defining democracy in this way. Firstly, we would claim that we have constructed a tightly unified concept from the perspective of which real-world political systems can be categorised in an unambiguous manner along many of the major dimensions of political power that have been of central concern within the democratic tradition. Secondly, we
13 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
would claim that this conception of the fully democratic community as a community of political equals points us clearly towards those socio-political values that have themselves constituted the main basis of the commitment to democracy in the modern world. Looking first at an expansion of this unified concept of democracy, we would claim that thinking of democracy as identical participation in the exercise of political power by all those subject to that power allows us to identify three distinct but logically inter-related dimensions that we refer to as the quantitative dimension, the procedural qualitative dimension and the dimension of real effectiveness. The quantitative dimension, along which is measured the first and most basic element of the distribution of political power, consists in the numerical ratio between the number of those with rights of participation in political power and the number of those systematically subject to the exercise of that power; in modern states it is the degree of inclusiveness of the franchise. The almost self-evident point here is that as the degree of inclusiveness increases, the more democratic, in this dimension, is the distribution of political power, up to the limit of maximally democratic enfranchisement, where everyone subject to the collectively binding decisions has the right to participate in the making of those decisions. Fundamentally important as enfranchisement is, the measure of inclusiveness is only a measure of the most basic aspect of democracy. To characterise the democratic quality of a political community we need to examine the procedures of political decision- making, we need to give an account of what type of participation enfranchisement entitles people to. In particular we need to examine the quality and equality of people’s democratic rights as defined by the procedures of collective decision-making. To understand what is at issue here, we need to specify the four logically distinct moments in any decision-making process. The first moment of any such process is the identification of the options to be decided upon; in formal decision-making terms, the drawing up of the agenda. Secondly, the decision-makers must comparatively assess the available options, such assessment leading to the third and central stage which consists in the selection of one of the options as the one to be implemented. The final moment, to which, of course, the other stages are preliminary, is the moment of implementation, the enactment of the decision arrived at. The procedural requirements of political equality, of maximally democratic decision-making, have implications for each of these
14 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
moments. Though democracy is sometimes popularly defined as one person, one vote; one vote, one weight, it is evident that people having equally weighted votes in a decision, the agenda for which they did not set, are not full participants in the exercise of power. Prisoners condemned to death, who are allowed to choose the form of their execution, do not have equal power with their judges and executioners. For decision- making to be fully democratic, the agenda must reflect whatever anyone within the community might consider a possibly desirable option. This, as we will see later, has quite strong implications for the democratic assessment of voting and electoral systems. At the second moment of decision-making, full political equality necessitates not what we would call specific decision-making procedures but the rights of free and equal access to relevant information and the equal rights to freely propound what anyone considers to be relevant arguments and conclusions. If, for example, some people were denied the right to put forward their particular beliefs and opinions on some matter to be decided, this would undoubtedly constitute a denial of equal rights in the determination of the outcome of the decision-making process. It is at the third stage, however, that institutional procedural matters come to the forefront, and that we encounter for the first time directly the central concern of this book. When an individual person is making a decision, the third-moment selection of the option to be implemented is simply the conclusion of the second moment of comparative assessment. Evidently, however, matters are completely different when a decision is to be arrived at “à la pluralité des voix” as the Marquis de Condorcet put it in his 18 th -century treatise on voting procedures. Much as we might think it desirable that collectively binding decisions should be based on full rational consensus emerging from free and open public deliberation, no decision-making procedure can assume that such consensus will be achievable; the problem of disagreement, of the plurality of different opinions is inescapable. And from our perspective, that problem is how to construct maximally democratic voting procedures, how to construct voting procedures that will most realise our democratic goals and ideals. The conventional wisdom that claims that equality in voting procedures requires one person, one vote; one vote, one weight, is true as far as it goes. If some persons do not have votes at all, then compared to those who do have votes and whose votes determine the outcomes, the voteless are obviously denied a share in formal power. And if, among those who do have votes, some people’s votes
15 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
count for, say, ten times the votes of others, a serious deviation from equality of power exists. The belief that once this much has been achieved there are no further democratic questions to be raised about voting systems is, however, seriously flawed. The rest of the book will, of course, demonstrate this at length, but to illustrate the seriousness of what is at stake, consider the following kind of situation, which is only a slight exaggeration of what actually happens quite often in real-world politics. Imagine a country in which there were only two political parties; let us assume that no more than two parties were really needed because political opinion right across the population was divided in the support of just these parties, with, say, 56% supporting party A and 44% supporting party B. Suppose further that the electoral system had certain obviously democratic features; universal suffrage, constituencies of equal size, anonymity of votes with each vote carrying the same weight, and so on. Add, however, just three further conditions; firstly, that the constituencies each elected a single representative, that the voting system was a straightforward simple majority one, and finally, the proportional support for the parties in each constituency was within a couple of percentage points of the national average. The inevitable result would be that party A would win 100% of the seats in the national legislature. Even abstracting from the likely consequences of that result in the ongoing policies of that country it is, as it were, mathematically demonstrable that the 44% of votes for party B candidates had no effect in determining the composition of the legislature and, hence, that each supporter of party B had, in terms of actually effective power, not only less power than a supporter of party A, but no power at all . Were such a situation to remain permanent in such a country, it seems obvious that the political system, in particular the voting system, is far from maximally democratic. What other kinds of system there are, and what can be said about them from the perspective of maximising democracy, will be the subject matter of the rest of this book. From the point of view of voting and electoral systems, democratic goals and ideals have most relevance to the first and second moments of collective decision-making, if we assume that the voting systems in question are maximally inclusive. To complete our account of strong democracy as full political equality, we need to look briefly at the fourth moment of decision-making, and the third dimension of political equality generally, the dimension of real equal effectiveness. The fourth moment of decision-
16 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
making is the moment of implementation, and the single democratic requirement at this stage is that the decision democratically arrived at can be implemented, as distinct from being subverted by some countervailing non-democratic power in the society. Crucial as this aspect of democratic political power is, its realisation does not depend upon internal procedural arrangements. The same is true of the whole third dimension of the real effectiveness of equal democratic rights. There are numerous examples where a group of people have formally equal rights but are, for one reason or another, unable to fully utilise those rights, with the result that in the actual participation in collective decision-making such groups are under-represented. The situation of women in almost all “advanced” democracies is a case in point. It is important from a theoretical point of view to state absolutely clearly that institutionally inscribed equal rights of participation only lead to democratic equality in practice if everyone has access to those resources necessary for the full utilisation of their rights. In measuring democratic achievement real equality of power is critical, alongside the democratic character of the decision procedures themselves. This conception of democracy, differentiated now into six dimensions – the quantitative dimension, the four dimensions of procedural equality, and the dimension of real effectiveness – forms, we would argue, an internally unified concept, with each dimension intelligibly contributing to overall political equality, to the Rousseauian identity of sovereign and subject. Our second reason for adopting this definition, in addition to analytic clarity, is that this understanding of democracy highlights the kinds of values and ideals that have formed the foundation of the normative commitment to democracy in the modern world. This argument can be elaborated by examining these ideals, which examination will also provide us with a more robust basis for a democratic assessment of voting procedures in the remainder of the work. To elucidate these values we might follow Robert A. Dahl in his identification of three basic principles ( Democracy and its Critics , pp. 83-105) the principle of equal intrinsic worth, the principle of strong equality and the principle of autonomy. While not on its own grounding the values of democracy, the principle of equal intrinsic worth is the foundation of all else, consisting as it does of the normative commitment that every person’s interests deserve equal consideration, that there is no person or group who can claim by right that their interests, their well-being, should take priority over the well-
17 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
being of others. There are various philosophical formulations suggested for this basic thesis of equal intrinsic worth (and there are numerous arguments concerning what it should imply in practice) but whatever we might believe about their rational conclusiveness, there is little doubt that mainstream ideological debate in the modern world all takes place on what Ronald Dworkin has called “the egalitarian plateau”. Outside a few extremist circles, for the purposes of public normative debate the equal importance, other things being equal, of the well-being of each person is a bedrock principle. It is evident, however, that this does not itself ground the unique superiority of democracy. Perhaps the best way of ensuring that set of political decisions that would underpin maximal equal well-being in a community would be to allocate political power to an elite minority of superior knowledge and devotion to the public interest, much as Plato had argued some 2,500 years ago. To argue from the egalitarianism of equal worth to democracy, that first principle must be backed up by something like Dahl’s “strong principle of equality” and the principle of autonomy. Unlike the equal intrinsic worth principle, which is a normative commitment, “strong equality” is a substantive empirical claim, which as such could be open to disproof. What is being asserted is that, in general, people “are adequately qualified to govern themselves” (Dahl, p. 97) or, to put the matter negatively, there is, at the end of the day, no other group better qualified to pursue people’s interests. As always, Dahl’s own arguments for this claim are commonsensical and plausible. The thesis that people’s interests in the political sphere would be better served if political power were vested not in the people but in some particular minority would itself require the truth of two assumptions to support it. The first is the knowledge assumption, the second is the assumption of integrity and benevolence of disposition. The knowledge needed to act in accordance with someone’s interests consists in knowledge of basic general goals and knowledge of the best means of achieving those goals. With respect to the former, Dahl claims that from a modern liberal perspective the determination of ultimate goals is not a matter of provable knowledge that one person might possess in greater measure, hence having the right to prescribe to others. In the matter of means, expertise may be relevant, but even here it is hard to find an area in which experts unanimously agree and expertise, anyway, can be employed by means of advisers, without giving the experts permanent irrevocable powers. When it comes to benevolence of disposition, the
18 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
democratic case is, if anything, stronger. Even if there was a wise and knowledgeable elite that always knew better that the people themselves what was in the people’s best interest, could we ever be certain that power vested in that elite would always be used for the general good and not for the benefit of the elite? An elite might be of a benevolent disposition, but then again they might not be. Whereas people might be foolish, but there is no systematically based orientation in people to be unconcerned with their own interests. On the contrary, whatever small basis there might be in people to be concerned with the well-being of others, it is a safe assumption that one is always concerned with one’s own well-being. Consequently, while temporary revocable delegation of power might be desirable, the total alienation of power to a minority on the assumption that that minority will remain in perpetuity benevolently disposed to the people’s interests is, to put it mildly, unsafe. The claim that, even were it safe, such monopolisation of power by some over others would still be undesirable and illegitimate, is what lies at the heart of the third principle, the principle of autonomy. Our discussion of autonomy goes beyond Dahl’s by identifying two distinct though related principles grounded in two types of autonomy, moral and de facto. The assertion of moral autonomy is a fundamental normative claim concerning what constitutes moral agency as such. As a principle it came to play a central role in how we think of the moral dimension of the person in the context of post- Reformation religious disputes, and is popularly referred to as “freedom of conscience”. To put the principle negatively, it rejects the right of an external agency to prescribe right and wrong to an individual conscience. Ultimately, a moral agent can only be bound by what that agent believes her/his duty to be. While coerced conformity to a rule might produce beneficial consequences, it is empty of moral value to the coerced person. For a person to be doing right, they must be doing what they believe is right. From the normative perspective of moral autonomy, people themselves are the only legitimate source of the prescription of right and wrong, of ultimate goals. In contrast, de facto autonomy is simply the power one has to control for oneself the circumstances of one’s life. It consists in a person’s actual ability to decide for him/herself, to choose in accordance with those decisions and to implement those choices. De facto autonomy is the actual power of self-determination. Moral autonomy is related to de facto autonomy as one of the important bases of the latter’s desirability.
19 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
If a person cannot be legitimately prescribed for by others, it would be better, other things being equal, for the person to be able to live in accordance with that moral autonomy by actually having the power to prescribe for her/himself and to be able to live in accordance with those prescriptions. Actual autonomy is, of course, never absolute (literal absolute autonomy would require literal omnipotence) and the autonomy of one person will limit the autonomy of others and vice versa. So, from an egalitarian perspective, the goal would be the maximisation of the autonomy of each as compatible with the autonomy of others. We could identify three basic reasons for valuing autonomy, the first of these being the straightforward pragmatic reason relating back to the type of argument Dahl used for his strong equality principle; other things being equal, one’s life is best left in one’s own hands. Further than this, however, our normative conception of what it is to be a mature person is based on the value of independence and responsibility for oneself. Finally, there is the conformity mentioned above of de facto autonomy with the principle of moral autonomy itself. Unsurprisingly, these three types of reason re-emerge as three types of reason justifying democratic principles of political power. Given that these three bases for a positive evaluation of democracy have crucial implications for the democratic assessment of voting procedures in particular, it is worth spelling them out in some detail. The first and most basic dimension of democracy consists in the direct consequences of democracy constituted by the flow of decisions and political outcomes. Fairly obviously, one of the most important things about a method of making decisions is the quality of decisions arrived at. While there may be many criteria for judging the rightness or the quality of decisions, there are two factors that are of unquestionable importance. Firstly, for any particular person or group a good decision will, all other things being equal, be a decision that furthers their interests, that has a positive impact on their well-being. As we said when discussing Dahl’s strong equality principle, while it should be emphasised that people are not necessarily infallible guides to what will best promote their interests, if individuals or groups are not represented in a decision- making process, there is no guarantee at all that their interests will be taken into consideration. Secondly, as Dahl in particular has argued, (D&C) irrespective of the specific content of decisions, there is a generic benefit deriving from knowing that one is living according to laws in the making of which one has oneself participated. Given
20 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
that in any political community there is likely to be disagreement and conflict of opinion over what should be done in any particular policy area, it follows that not everyone can get all of what they want all of the time. With regard to the specific content of decisions and the generic autonomy in living according to laws of one’s own choosing, the general principle of democratic equality should be the maximisation of equal effectiveness in the determination of political outcomes. What level of equality of effectiveness can be achieved, in the light of possibly quite radical disagreement, will be one of the central problems we will deal with in our analysis of different preference aggregation procedures. The general principle of the desirability of equal effectiveness is, however, clear. Many supporters of democracy, from Rousseau and J.S. Mill to the twentieth-century theorists of participatory and deliberative democracy, have argued that political systems should be evaluated in terms of their indirect consequences, and not just as arrangements for arriving at specific decisions. Particular emphasis is placed on the beneficial consequences that can stem from informed participation in collective decision-making. One idea is that, even if a person disagrees with certain specific decisions, there are general benefits to be gained from living in a community in which decisions are made on the basis of open public debate that is informed by the diversity of all perspectives relevant to the issues being decided. Secondly, at an individual level it is claimed that active participation in collective decision-making can be the basis of a multi-faceted personal development. Participation can lead people to become more informed about public affairs, it can stimulate a sympathetic appreciation of alternative perspectives on issues, which might itself provide a foundation for a civic spirit that evaluates things from a community rather than a narrow partisan point of view. Finally, it is argued that involvement can be psychologically empowering, enhancing both a sense of the person’s own worth and a feeling of communal belonging. It is important to note that the generic and the personal benefits of informed participation are critically dependent on the level and quality of actual participation and while these, no doubt, depend on many factors, it will be part of our argument that they might also be related to institutional factors such as the nature of voting and electoral systems. The final dimension of democracy relevant to its positive normative assessment is not concerned with consequences at all, but focuses on features intrinsic to democratic
21 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
systems as such. A democratic system is constituted by a set of rights, a set of entitlements to participate in political decision-making, and it can be argued that simply being so entitled is something of intrinsic worth in itself. To begin with, the constitutional inscription of democratic rights is a recognition of the fundamental moral autonomy of the person, it is a public recognition that there is no agency other than people themselves that can claim by right the authority to command. Secondly, the political equality of full democracy is a public acclamation of the equal worth of all members of the community as human beings. It can be very strongly argued that recognition of one’s equal worth as a responsible human being is something of intrinsic value, as compared to being classified as a second-class citizen with less than full, human status. It is no doubt true that these very significant high-level values of democracy are primarily dependent on maximal enfranchisement and basic equality of voting rights, rather than on the details of preference aggregation procedures. There are some situations however, in which, for example, permanent minorities might find themselves without significant effectiveness and, despite basic equality of democratic rights, feel themselves effectively disenfranchised by reason of the technicalities of a voting and electoral system. From that perspective it might be argued that recognition of equal moral autonomy and the symbolic benefits of public ratification of equal worth do have implications for the choice of decision-making procedure. The concern of the rest of this work will be to investigate the ways that the different types of voting and electoral procedure impact on the achievement of democratic values. It should not be thought, however, that through our logical and empirical analysis of voting procedures we are going to discover some ideal method of collective decision-making that will solve the problems of political conflict at one fell swoop. In the first place, though some procedures might be clearly superior to others in certain respects, there is probably no single set of procedures that is ideal in every way. Secondly, a just, effective and democratic method of government requires a lot more than procedures. A procedurally fair aggregation of raw preferences that are themselves, say, ill-informed and bigoted, is not going to result in perfect justice. Even from the perspective of democratic equality, a deliberative process through which preferences might become more informed, rational and impartial would be more desirable than hasty decisions that were thought to have complete democratic legitimacy simply because the procedure used was a fair one.
22 Voting Systems: Chapter One Eddie Hyland Democracy, democratic ideals and voting
Finally, as we emphasised in our general account of the nature of democracy, procedures are of little use if some people do not have the resources needed to utilise the procedures. However, though procedures on their own can never be a sufficient condition of ideally democratic decision-making, they can, as the rest of this book will demonstrate, contribute in a very significant way to the achievement of democratic ideals.
23